The probability of default is one of the most important risk parameters estimated in credit institutions, especially banks, and plays a major role in credit risk analysis and management. Given the fact that one of the fundamental activities of banks is granting loans, the banking industry places a great deal of emphasis on credit risk. Credit risk is commonly understood as the potential that a borrower or counterparty will fail to meet its contractual obligations. For a bank, it is crucial to evaluate the credit risk associated with potential clients, as well as with actual clients. Credit risk evaluation of clients is performed within credit scoring, which is a process for predicting the probability that a loan applicant or a client will default.
The expected credit loss (ECL) or impairment calculation rules set by the IFRS9 standard require the financial institutions to calculate expected loss conditional on macroeconomic factors on a point-in-time basis.
Let us briefly understand requirements of probability of default from IFRS-9 point of view:
- PD estimates should be unbiased (“best estimate PD”), i.e. PD should accurately predict number of defaults and does not include optimism or conservatism. Wherever, BASEL models (also known as regulatory or capital models) are used as a starting point to calculate ECL, appropriate adjustments need to be made to remove inherent conservatism.
- Estimated PDs should be point-in-time, i.e., adjusted, where necessary, to reflect the effects of the current economic conditions.
- The PD estimates should be recalibrated on a regular basis, or monitoring should be provided to show why recalibration was not necessary.
- The PDs should be calculated with appropriate segmentation – the bank should consider risk drivers in respect of borrower risk, transaction risk and delinquency status in assigning exposures to PD model segments.
The IFRS-9 introduces the concept of staging i.e. each of the loans would be categorized into different stages depending upon the credit deterioration of the account since initial recognition. Under the general approach, an entity must determine whether the financial asset is in one of three below mentioned stages in order to determine both the amount of ECL to recognize as well as how interest income should be recognized.
Stage 1: Low Risk
Stage 2 – Loan’s credit risk has increased significantly since initial recognition.
Stage 3 – If the loan’s credit risk increases to the point where it is considered credit-impaired.
Depending upon the stage in which a particular account falls into either 12 month PD or lifetime PD should be calculated.
- 12-month ECLs (Stage 1): Applies to all items (from initial recognition) as long as there is no significant deterioration in credit quality •
- Lifetime ECLs (Stages 2 and 3): Applies when a significant increase in credit risk has occurred on an individual or collective basis
12 Month PD: Though existing BASEL models could be used to calculate 12 months IFRS-9 PD models as default definition both for BASEL and IFRS-9 is same (90 DPD, Forbearance and Charge Off ) but there are certain inherent differences in assumptions for BASEL and IFRS-9.
- BASEL models can be TTC, PIT, or hybrid with an additional requirement (in CRD4) that capital should not be procyclical, hence acceptable PD models and rating systems that produce PD parameters for capital calculations tend to be TTC or hybrid. Whereas IFRS-9 models are PiT models and requires inclusion of Micro Economic factor into the model.
- The second difference between Basel and IFRS9 requirements is the IFRS9 requirement for PDs to be unbiased, or “best estimate”, which contrasts to the Basel requirement for PDs to be “conservative”.
If credit risk of a customer is low at the reporting date, management can measure impairment using 12-month ECL, and so it does not have to assess whether a significant increase in credit risk has occurred.
In order for this operational simplification to apply, the customer has to meet the following requirements:
- The borrower is considered, in the short term, to have a strong capacity to meet its obligations.
- The lender expects, in the longer term that adverse changes in economic and business conditions might, but will not necessarily; reduce the ability of the borrower to fulfil its obligations.
There are certain cases where using 12 Month ECL as an approximation for Lifetime ECL is not suitable like, the bullet repayment loans where the payment obligations of the debtor are not significant during the first 12 months of the loan facility or for certain financial instruments changes in credit-related factors only have an impact beyond 12 months.
12 Month PD: Basel models which could be used as a base for 12 month IFRS-9 models are generally “Logistic Regression” models. Where Probability of Default is calculated based on certain behavioral variables.
Next section would cover Lifetime PD for IFRS-9.
To better understand IFRS-9 it is advised to start from the first section (Introduction IFRS-9).